Gradian

Gradian

Find the data that broke your fine-tune.

13 followers

Training-data attribution for LoRA fine-tuned LLMs. Gradian scores every training example against the capability you lost, then runs a battery of dataset and config checks for the failures influence math misses.
Gradian gallery image
Gradian gallery image
Gradian gallery image
Free
Launch Team
AWS Activate Credits
AWS Activate CreditsCredits that fuel your startup’s next milestone
Promoted